Assessment and Criticality of Defects and
Damage in Adhesively Bonded Composite
Structures
W R Broughton, M R L Gower and M J Lodeiro
National Physical Laboratory, Teddington, Middlesex, TW11 0LW
Workshop on NDT/SHM Requirements for Aerospace Composites
National Composites Centre, 9-10 February 2016
Agenda
Introduction
Requirements of SHM/NDE
Limitations of conventional NDE
Production and in-service defects/damage
Failure mechanisms
Adhesively bonded joints
Sandwich structures
Kissing bonds
SHM/NDE techniques
Defect criticality framework
Conclusions
Source: Airbus website
Requirements of SHM/NDE Techniques
Currently, there is a lack of design methodologies, reliable NDEtechniques and useable data for adhesively bonded composite andmetallic structures/systems
Requirement for in-situ, real-time SHM techniques for accuratemonitoring and quantification of deformation and damage (i.e.improved probability of detection (PoD) of safety critical defects), andremnant life of bonded composite and hybrid engineering structuresfor in-service performance assessment
Improved predictive modelling of failure mechanisms (i.e. damageinitiation and growth) in adhesively bonded composites andcomposite sandwich constructions under complex loading conditions(including static, cyclic fatigue and hostile environments) – fracturemechanics and stress-based failure criteria approaches
Reliable techniques for simulating (reference defect artefacts (RDAs)),detecting and characterising safety critical defects in adhesivelybonded composite structures
Limitations of Conventional NDE Techniques
Inability to detect small defects before they grow to a critical size
Inspection of structural parts with complex geometries (i.e. bonded
joints)
Unable to dismantle critical parts in bonded structures for detailed
inspection
Limitations on technique efficiency and reliability for accurate
localisation and detailed characterisation (shape/dimensions) of
certain types of damage (i.e. kissing bonds/kissing de-bonds)
Difficulty in detecting certain defects extends to the inspection of
adhesively bonded repairs of composites
NDE data - generally there is no direct feedback to in-situ, real-time
engineering design software for rapid structural integrity
monitoring and assessment (i.e. decision making software)
Production/In-service Defects/Damage
in Composite LaminatesThermal Effects
Non-uniform cure (thick sections) - poor consolidation
Poor cure and heat damage
Exothermic chemical reactions (thick sections)
Residual stresses microcracking + delaminations
Delaminations
Fibre Related Defects
Fibre breakage/fraying - inner radii of curved structures
Fibre wash (or displaced fibres) or whorls
Poor fibre alignment and incorrect stacking sequence
Warpage/fibre folding (pultrusions)
Wrinkling/kinking (i.e. out-of-plane deformation)
Matrix (Resin) Related Defects
Microcracking
“Unwetted” (resin starved or dry) areas
Resin rich areas
Voids (or porosity)
Mechanical Handling/Processing/Machining Induced Defects
Inclusions (e.g. release film, chemical contaminants)
Poor ply abutment
Local buckling or bulging
Steps (thickness variations)
Fibre wrinkling/kinking and surface rippling
Dents, nicks, gouges and scratches
Production/In-service Defects/Damage
in Composite Laminates
In-service Damage
Fibre fracture and pull-out
Impact damage – transverse cracking/delaminations
Lightning strike – thermal and mechanical damage
Moisture/chemical ingress
Elevated and sub-zero temperatures
Cyclic fatigue and creep rupture
Corrosion/erosion (material thinning)
Fire damage
Sandwich Structures
Sandwich skin-to-core de-bonding
Crushed sandwich core
Production/In-service Defects/Damage
in Composite Laminates
Fibre Disbonds – interfacial failure between adherend and adhesive or
the surface treatment (e.g. primer)
Zero-volume disbond (kissing bond) – interface is bonded, but bond
strength is not assured
Poor cure poor cohesive strength resulting from either poor mixing,
inadequate temperature control, light or other form of energy to
activate cure, or pressure during the cure cycle
Porosity and voids – due to volatiles (e.g. water vapour) within the
adhesive, entrapped air or insufficient application of adhesive
Cracking within adhesive - due to incorrect cure or brittleness of cured
resin (brittle resins are susceptible to cracking under impact loading
and thermal cycling)
Residual stresses in adhesive layer due to differences in the coefficient
of thermal expansion (CTE) between the adherends and adhesive
resulting from processing at elevated temperatures
FRP delamination – occurs when the interfacial strength at the
adhesive/adherend regions is stronger than the fibre/matrix interface
Production/In-service Defects/Damage
in Bonded Joints
Production/In-service Defects/Damage
in Bonded Joints
Adherend
Adherend
Adhesive PorosityVoid
Zero Volume (Kissing) Bond De-bond
Cracks
Adherend
Metal (titanium/aluminium) tensile yielding - in-plane/bending loads
Composite tensile failure/rupture - in-plane/bending loads
Composite delamination - interlaminar shear
Composite transverse tensile stress (Poisson’s effect for 0 plies)
Interface
Interface - shear or peel stresses
Adhesive Layer
Cohesive failure - shear and peel stresses
Metal tensile yielding
Failure in Adhesively Bonded Joints
Composite
Adhesive
Metal
Tensile failure/rupture
Composite/adhesive interface de-bonding
Adhesive layer (cohesive) failure
Metal and adhesive interface de-bonding
Delamination
Production/In-service Defects/Damage
in Sandwich Construction
Zero Volume (Kissing) Bond De-bond
Cracks
Top FRP Skin
Bond-line
Bottom FRP Skin
Butt-jointDe-bond
Skin/core de-bond
Flaws and voidsShear crack
Damage is difficult to detect due the multi-
layered structure of sandwich construction;
especially if access is limited to one surface
In-service Damage of Sandwich
Constructions
Impact delaminations in top skin of
sandwich construction
Skin-to-core de-bond in GFRP/PU
foam sandwich construction
Core crushing of CFRP/Nomex
construction
Failure Modes in Sandwich
Constructions
Facing Transverse shear Local crushing
Panel Buckling Shear Crimping Face Wrinkling
Intracell Buckling (Dimpling)
Source: Hexcel website
Kissing Bonds
Delaminations Kissing Bond
Separation between FRP plies, zero strength Mechanical/frictional contact without chemical bond
Characteristics of a weak bond (defective adhesive bond) are defined
below:
Strength as measured by mechanical testing is below 20% of the
nominal bond strength,
Mode of failure must be adhesive in type (i.e. purely at the
adherend/adhesive interface), and
Undetectable from normal bonds with conventional NDE
techniques – includes kissing bonds (weak adhesive bond)
produced through surface contamination with release agent
Kissing Bond – Simulation Using Different Surface Pre-treatments in CFRP Laminates
Surface Treatment* Failure Stress (MPa)
N°. Detail “As Received” Surface Pre-treatment Grit Blasted Pre-treatment
1 Tygacote® (release agent) failed removing from bonding alignment rig 2.74
2 Graphite (powder spray) 1.84 1.73
3 Beeswax (release agent) 9.92 12.60
4 Silicone (release agent) 8.76 14.14
5 PTFE (dry lubricant spray) 6.17 1.83
* - 4 coats (layers) applied in each case
Note: Non-defective material failed in one of the parent laminate at 42 MPa, indicating that the adhesive
bond strength was superior to the interlaminar (through-thickness) tensile strength of the laminate
SHM/NDE Techniques for Inspection
of Adhesively Bonded JointsNDE Techniques
Visual inspection (CCTV cameras, endoscopes)
Tap test (coin, Woodpecker)
Ultrasonics (contact, immersion)
C-scan
Scanning acoustic microscopy (SAM): 5-400 MHz
Non-linear Elastic Wave Spectroscopy (NEWS): 0-500 KHz
X-radiography and X-ray computed tomography
Pulse thermography (including thermal shock)
Microwave
Eddy current (metallic and CFRP systems)
Laser shearography
SHM Techniques
Acoustic emission
Strain gauges
Fibre Bragg grating (chirped FBGs) sensors
Digital image correlation (DIC)
Electrical self-sensing
Single Frequency NEWS
Reference Defect Artefacts (RDAs)
Microwave of thick bond-line
containing artificial defects
(24 GHz)
1MHz thru-transmission
ultrasonic c-scan
Manufacture of Reference
Defect Artefacts (RDAs)
20
EMRP JRP ENG57
Validated Inspection Techniques For
Composites In Energy Applications
(VITCEA)
July 2014 – June 2017
European Metrology Research
Programme (EMRP) – EURAMET/NMS
(EMRP Call: Energy 2013)
VITCEA Project Structure
Design and
manufacture of
reference and
natural defect
artefacts
(RDAs & NDAs)
Manufacture &
characterisation
of reference
materials used
in RDAs &
NDAs
Practical application and experimental
optimisation of techniques
Simulation/modelling capability
development (except shearography)
Scanning Techniques
• Phased array and air-coupled
ultrasonics
• Microwave
Full-Field Techniques
• Active thermography
• Laser shearography
Inter-
comparison
exercises,
field trials
and
finalisation of
procedures
http://projects.npl.co.uk/vitcea/
Defects - Probability of Occurrence
and Impact on Structural Integrity
DefectProbability of
Occurrence
Impact on Structural
Integrity
Adherend Surface Contamination Low-Medium Severe
Delaminations/Debonds Medium Severe
Partial/Local Cure Low Moderate
Inclusions Low Severe
Voids/Porosity Moderate Medium
Residual Stresses/Thermal Cracking Low-Medium Moderate
Non-uniform Adhesive Distribution* High Moderate
Crushed Sandwich Core Low Severe
Sandwich Skin/Core De-bonding Low-Medium Severe
* Non-uniform bond-line - resin rich and depleted regions
Three-Level Approach to Assess
Damage in Structures
NPL Measurement Good Practice Guide No. 78 “Assessment
and Criticality of Defects and Damage in Material Systems”,
M Gower, G Sims R Lee, S Frost, M Stone and M Wall
The National Measurement System is the UK’s national infrastructure of measurement
Laboratories, which deliver world-class measurement science and technology through four
National Measurement Institutes (NMIs): LGC, NPL the National Physical Laboratory, TUV NEL
The former National Engineering Laboratory, and the National Measurement Office (NMO).
The National Measurement System delivers world-class
measurement science & technology through these organisations
24